- 01 Feb, 2013 7 commits
-
-
Chris Mason authored
The extent buffers have a refs_lock which we use to make coordinate freeing the extent buffer with operations on the radix tree. On tree roots and other extent buffers that very cache hot, this can be highly contended. These are also the extent buffers that are basically pinned in memory. This commit adds code to cmpxchg our way through the ref modifications, and as long as the result of the reference change is still pinned in ram, we skip the expensive spinlock. Signed-off-by:Chris Mason <chris.mason@fusionio.com>
-
Chris Mason authored
With the new raid56 code, we want to make sure we're properly aligning our allocation clusters with -o ssd Signed-off-by: -
Chris Mason authored
Buffered writes and DIRECT_IO writes will often break up big contiguous changes to the file into sub-stripe writes. This adds a plugging callback to gather those smaller writes full stripe writes. Example on flash: fio job to do 64K writes in batches of 3 (which makes a full stripe): With plugging: 450MB/s Without plugging: 220MB/s Signed-off-by: -
Chris Mason authored
The stripe cache allows us to avoid extra read/modify/write cycles by caching the pages we read off the disk. Pages are cached when: * They are read in during a read/modify/write cycle * They are written during a read/modify/write cycle * They are involved in a parity rebuild Pages are not cached if we're doing a full stripe write. We're assuming that a full stripe write won't be followed by another partial stripe write any time soon. This provides a substantial boost in performance for workloads that synchronously modify adjacent offsets in the file, and for the parity rebuild use case in general. The size of the stripe cache isn't tunable (yet) and is set at 1024 entries. Example on flash: dd if=/dev/zero of=/mnt/xxx bs=4K oflag=direct Without the stripe cache -- 2.1MB/s With the stripe cache 21MB/s Signed-off-by: -
David Woodhouse authored
This builds on David Woodhouse's original Btrfs raid5/6 implementation. The code has changed quite a bit, blame Chris Mason for any bugs. Read/modify/write is done after the higher levels of the filesystem have prepared a given bio. This means the higher layers are not responsible for building full stripes, and they don't need to query for the topology of the extents that may get allocated during delayed allocation runs. It also means different files can easily share the same stripe. But, it does expose us to incorrect parity if we crash or lose power while doing a read/modify/write cycle. This will be addressed in a later commit. Scrub is unable to repair crc errors on raid5/6 chunks. Discard does not work on raid5/6 (yet) The stripe size is fixed at 64KiB per disk. This will be tunable in a later commit. Signed-off-by: -
David Woodhouse authored
We'll want to merge writes so they can fill a full RAID[56] stripe, but not necessarily reads. Signed-off-by:
David Woodhouse <David.Woodhouse@intel.com> Signed-off-by:
-
Eric Sandeen authored
If we remove a missing device, bdev is null, and if we send that off to btrfs_kobject_uevent we'll panic. Signed-off-by:
Eric Sandeen <sandeen@redhat.com> Signed-off-by:
Josef Bacik <jbacik@fusionio.com> Signed-off-by:
-
- 19 Dec, 2012 1 commit
-
-
Chris Mason authored
This reverts commit 6a7a665d. This was bug was fixed differently in 3.6, so this commit isn't needed. Conflicts: fs/btrfs/ctree.c Signed-off-by:
-
- 18 Dec, 2012 1 commit
-
-
Chris Mason authored
This reverts commit 95c80bb1. The bug addressed by this commit was fixed differently back in 3.6 Signed-off-by:
-
- 17 Dec, 2012 31 commits
-
-
Liu Bo authored
Users report a bug, the reproducer is: $ mkfs.btrfs /dev/loop0 $ mount /dev/loop0 /mnt/btrfs/ $ mkdir /mnt/btrfs/dir $ chattr +C /mnt/btrfs/dir/ $ dd if=/dev/zero of=/mnt/btrfs/dir/foo bs=4K count=10; $ lsattr /mnt/btrfs/dir/foo ---------------C- /mnt/btrfs/dir/foo $ filefrag /mnt/btrfs/dir/foo /mnt/btrfs/dir/foo: 1 extent found ---> an extent $ dd if=/dev/zero of=/mnt/btrfs/dir/foo bs=4K count=1 seek=5 conv=notrunc,nocreat; sync $ filefrag /mnt/btrfs/dir/foo /mnt/btrfs/dir/foo: 3 extents found ---> with nocow, btrfs breaks the extent into three parts The new created file should not only inherit the NODATACOW flag, but also honor NODATASUM flag, because we must do COW on a file extent with checksum. Signed-off-by:
Liu Bo <bo.li.liu@oracle.com> Signed-off-by:
-
Chris Mason authored
The handling for directory crc hash overflows was fairly obscure, split_leaf returns EOVERFLOW when we try to extend the item and that is supposed to bubble up to userland. For a while it did so, but along the way we added better handling of errors and forced the FS readonly if we hit IO errors during the directory insertion. Along the way, we started testing only for EEXIST and the EOVERFLOW case was dropped. The end result is that we may force the FS readonly if we catch a directory hash bucket overflow. This fixes a few problem spots. First I add tests for EOVERFLOW in the places where we can safely just return the error up the chain. btrfs_rename is harder though, because it tries to insert the new directory item only after it has already unlinked anything the rename was going to overwrite. Rather than adding very complex logic, I added a helper to test for the hash overflow case early while it is still safe to bail out. Snapshot and subvolume creation had a similar problem, so they are using the new helper now too. Signed-off-by:
Pascal Junod <pascal@junod.info>
-
Josef Bacik authored
This confuses and angers lockdep even though it's ok. We don't really need the lock for free space inodes since only the transaction committer will be reserving space. Thanks, Signed-off-by:
-
Josef Bacik authored
This happens because writeback_inodes_sb_nr_if_idle does down_read. This doesn't work for us and it has not been fixed upstream yet, so do it ourselves and use that instead so we can stop having this stupid long standing lockup. Thanks, Signed-off-by:
-
Filipe Brandenburger authored
When a new file is created with btrfs_create(), the inode will initially be created with permissions 0666 and later on in btrfs_init_acl() it will be adapted to mask out the umask bits. The problem is that this change won't make it into the btrfs_inode unless there's another change to the inode (e.g. writing content changing the size or touching the file changing the mtime.) This fix adds a call to btrfs_update_inode() to btrfs_create() to make sure that the change will not get lost if the in-memory inode is flushed before other changes are made to the file. Signed-off-by:
Filipe Brandenburger <filbranden@google.com> Reviewed-by:
-
Liu Bo authored
Raid properties can be shared among raid calculation code, we can put them into a global table to keep it simple. Signed-off-by:
-
Stefan Behrens authored
This fixes a very special case that can be reproduced by just disconnecting a disk at runtime, and without unmounting the filesystem first, start scrub on the filesystem with the disconnected disk. All read and write EIOs are handled correctly, only the first superblock is an exception and gives a BUG() in a subfunction. The BUG() is correct, it would crash later otherwise. The subfunction must not be called for superblocks and this is what the fix changes. Reported-by:
Joeri Vanthienen <mail@joerivanthienen.be> Signed-off-by:
Stefan Behrens <sbehrens@giantdisaster.de> Signed-off-by:
-
Josef Bacik authored
This starts a transaction and dirties the inode everytime we call it, which is super expensive if you have a write heavy workload. We will be updating the inode when the IO completes and we reserve the space for the inode update when we reserve space for the write, so there is no chance of loss of information or enospc issues. Thanks, Signed-off-by:
-
Josef Bacik authored
I noticed while doing fsync tests that we were always dropping the path and re-searching when we first cow the log root even though we've already gotten the write lock on the root. That's because we don't take into account that there might not be a parent node, so fix the check to make sure there is actually a parent node before we undo all of this work for nothing. Thanks, Signed-off-by:
-
Josef Bacik authored
If we are syncing over and over the overhead of doing all those maps in fill_inode_item and log_changed_extents really starts to hurt, so use map tokens so we can avoid all the extra mapping. Since the token maps from our offset to the end of the page make sure to set the first thing in the item first so we really only do one map. Thanks, Signed-off-by:
-
Josef Bacik authored
This gets called at least 4 times for every level while adding an object, and it involves 3 kmapping calls, which on my box take about 5us a piece. So instead use a token, which brings us down to 1 kmap call and makes this function take 1/3 of the time per call. Thanks, Signed-off-by:
-
Josef Bacik authored
Our token logic depends on token->kaddr being set, and if it is not it sets everything properly as needed. So instead of memsetting just set token->kaddr to NULL. Thanks, Signed-off-by:
-
Josef Bacik authored
No reason to set the path blocking or loop through all of the pages if the extent buffer isn't actually marked dirty. Thanks, Signed-off-by:
-
Josef Bacik authored
This is a high traffic function, let's try and do as little as possible during normal operations shall we? Signed-off-by:
-
Josef Bacik authored
We don't really need to copy extents from the source tree since we have all of the information already available to us in the extent_map tree. So instead just write the extents straight to the log tree and don't bother to copy the extent items from the source tree. Signed-off-by:
-
Josef Bacik authored
You'd think path->keep_locks would keep all the locks wouldn't you? You'd be wrong. It only keeps them if the slot is pointing to the last item in the node. This is for use with btrfs_next_leaf, which needs this sort of thing. But the horrible horrible things I'm going to do to the tree log means I really need everything held from root to leaf so I can add and delete items in the same search. Thanks, Signed-off-by:
-
Josef Bacik authored
We are going to use EM's to log extents in the future, so we need to not mark them as prealloc if they aren't actually prealloc extents. Instead mark them with FILLING so we know to ammend mod_start/mod_len and that way we don't confuse the extent logging code. Thanks, Signed-off-by:
-
Josef Bacik authored
If we've written to a prealloc extent we need to know the original block len for the extent. We can't figure this out currently since ->block_len is just set to the extent length. So introduce ->orig_block_len so that we know how many bytes were in the original extent for proper extent logging that future patches will need. Thanks, Signed-off-by:
-
Josef Bacik authored
The tree logging stuff needs the csums to be on the ordered extents in order to log them properly, so mark that we're sync and inline the csum creation so we don't have to wait on the csumming to be done when logging extents that are still in flight. Thanks, Signed-off-by:
-
Josef Bacik authored
We don't copy inode items anwyay, we just copy them straight into the log from the in memory inode. So if we know we're only logging the inode, don't bother dropping anything, just try to insert it and either if it succeeds or we get EEXIST we can update the inode item in the log and carry on. Thanks, Signed-off-by:
-
Josef Bacik authored
Currently we copy all the file information into the log, inode item, the refs, xattrs etc. Except most of this doesn't change from fsync to fsync, just the inode item changes. So set a flag if an xattr changes or a link is added, and otherwise only log the inode item. Thanks, Signed-off-by:
-
Anand Jain authored
Originally root_times_lock was introduced as part of send/receive code however newly developed patch to label the subvol reused the same lock, so renaming it for a meaningful name. Signed-off-by:
Anand Jain <anand.jain@oracle.com> Signed-off-by:
-
Lukas Czerner authored
Currently udev does not know about the device being removed from the file system. This may result in the situation where we're unable to mount the file system by UUID or by LABEL because the by-uuid and by-label links may still point to the device which is no longer part of the btrfs file system and hence does not have any btrfs super block. It can be easily reproduced by the following: mkfs.btrfs -L bugfs /dev/loop[0-6] mount /dev/loop0 /mnt/test btrfs device delete /dev/loop0 /mnt/test umount /mnt/test mount LABEL=bugfs /mnt/test <---- this fails then see: ls -l /dev/disk/by-label/bugfs which will still point to the /dev/loop0 We did not noticed this before because libblkid would send the udev event for us when it notice that the link does not fit the reality, however it does not do that anymore and completely relies on udev information. Fix this by sending the KOBJ_CHANGE event to the bdev kobject after successful device removal. Note that this does not affect device addition, because we will open the device prior the addition from userspace and udev will notice that and reread the device afterwards. Signed-off-by:
Lukas Czerner <lczerner@redhat.com> Signed-off-by:
-
Miao Xie authored
ret variant may be set to 0 if we read page successfully, but it might be released before we lock it again. On this case, if we fail to allocate a new page, we will return 0, it is wrong, fix it. Signed-off-by:
Miao Xie <miaox@cn.fujitsu.com> Signed-off-by:
-
Miao Xie authored
Since we can pre-allocate the space past EOF, we should be able to reclaim that space if we need. This patch implements it by removing the EOF check. Though the manual of fallocate command says we can use truncate command to reclaim the pre-allocated space which past EOF, but because truncate command changes the file size, we must run several commands to reclaim the space if we don't want to change the file size, so it is not a good choice. Signed-off-by:
-
Miao Xie authored
Steps to reproduce: # mkfs.btrfs <disk> # mount <disk> <mnt> # dd if=/dev/zero of=<mnt>/<file> bs=512 seek=5 count=8 # fallocate -p -o 2048 -l 16384 <mnt>/<file> # dd if=/dev/zero of=<mnt>/<file> bs=4096 seek=3 count=8 conv=notrunc,nocreat # umount <mnt> # dmesg WARNING: at fs/btrfs/inode.c:7140 btrfs_destroy_inode+0x2eb/0x330 The reason is that we inputed a range which is beyond the end of the file. And because the end of this range was not page-aligned, we had to truncate the last page in this range, this operation is similar to a buffered file write. In other words, we reserved enough space and clear the data which was in the hole range on that page. But when we expanded that test file, write the data into the same page, we forgot that we have reserved enough space for the buffered write of that page because in most cases there is no page that is beyond the end of the file. As a result, we reserved the space twice. In fact, we needn't truncate the page if it is beyond the end of the file, just release the allocated space in that range. Fix the above problem by this way. Signed-off-by:
-
Miao Xie authored
(start + len) is the start of the adjacent extent, not the end of the current extent, so we should not use it to check the hole is on the same page or not. Signed-off-by:
-
Miao Xie authored
We forget to release the reserved space in the error path of delalloc reservatiom, fix it. Signed-off-by:
-
Miao Xie authored
If we runt the direct IO, we should not run auto defrag, because it may introduce buffered IO vs direcIO problem, and make direct IO slow down. Signed-off-by:
-
Liu Bo authored
Value 0 is not a tree id, so besides an upper limit, a lower limit is necessary as well while parsing root types of tracepoint. Signed-off-by:
-
Wang Sheng-Hui authored
We should use ctl->unit for free space calculation instead of block_group->sectorsize even though for free space use_bitmap or free space cluster we only have sectorsize assigned to ctl->unit currently. Also, we can keep it consisten in code style. Signed-off-by:
Wang Sheng-Hui <shhuiw@gmail.com> Signed-off-by:
-
